Package Basics

Getting started with polmineR (CLI And GUI usage).

polmineR()

polmineR-package

get_info() show_info()

Generic methods defined in the polmineR package

Corpora and subcorpora

Corpora, subcorpora/partitions, and partition_bundle objects are the point of departure for any analysis using polmineR. Constructors introduced in the next section require an understanding of basic methods and classes to generate and manage (sub-)corpora

use()

Add corpora in R data packages to session registry.

corpus(<character>) corpus(<missing>)

Corpus class initialization

name(<corpus>) get_corpus(<corpus>) show(<corpus>) `$`(<corpus>) get_info(<corpus>) show_info(<corpus>)

Corpus class methods

subset(<corpus>) subset(<character>) subset(<subcorpus>) subset(<remote_corpus>) subset(<subcorpus_bundle>)

Subsetting corpora and subcorpora

ocpu_exec()

Execute code on OpenCPU server

aggregate(<slice>)

Virtual class slice.

partition()

Initialize a partition.

partition_bundle()

Generate bundle of partitions.

p_attributes()

Get p-attributes.

s_attributes()

Get s-attributes.

is_nested()

Check whether s-attributes of corpus are nested

tree_structure()

Show the structure of s-attributes

as.speeches()

Split corpus or partition into speeches.

summary(<subcorpus>) `name<-`(<subcorpus>) get_corpus(<subcorpus>) size(<subcorpus>)

The S4 subcorpus class.

show(<subcorpus_bundle>) merge(<subcorpus_bundle>) merge(<subcorpus>) split(<subcorpus>) split(<corpus>) split(<subcorpus_bundle>)

Bundled subcorpora

regions() as.regions() as.data.table(<regions>)

Regions of a CWB corpus.

Methods I: Basic Vocabulary

Data analysis using polmineR will usually start by calling a constructor method resulting in an S4 class object with the name of the calling method. All methods basic vocabulary can be applied to corpora (defined by a character vector), to partitions, and to partition_bundle objects, some are applicable to additional classes. These methods are designed to be used in a pipe.

kwic()

Perform keyword-in-context (KWIC) analysis.

hits()

Get hits for query

ranges()

Get ranges for query.

count()

Get counts.

dispersion()

Dispersion of a query or multiple queries.

ngrams()

Get N-Grams

context()

Analyze context of a node word.

cooccurrences()

Get cooccurrence statistics.

features()

Get features by comparison.

as.phrases(<ngrams>) as.phrases(<matrix>) as.character(<phrases>) concatenate_phrases()

Manage and use phrases

Methods II: Processing Objects

The polmineR package offers a repertoire to modify base objects, and for type conversion. Methods for modyfing classes and type conversion.

size()

Get Number of Tokens.

terms(<partition>) terms(<subcorpus>) terms(<corpus>) terms(<character>)

Get terms in partition or corpus.

enrich()

Enrich an object.

trim() punctuation

Trim an object.

noise()

detect noise

encoding() `encoding<-`()

Get and set encoding.

as.TermDocumentMatrix() as.DocumentTermMatrix()

Generate TermDocumentMatrix / DocumentTermMatrix.

as.sparseMatrix()

Type conversion - get sparseMatrix.

as.VCorpus(<partition_bundle>)

Get VCorpus.

chisquare()

Perform chisquare-text.

ll()

Compute Log-likelihood Statistics.

t_test()

Perform t-test.

pmi()

Calculate Pointwise Mutual Information (PMI).

means()

calculate means

weigh()

Apply Weight to Matrix

mail() browse()

Defunct functionality

dotplot()

dotplot

name(<textstat>) name(<character>) `name<-`(<textstat>) round(<textstat>) sort(<textstat>) as.bundle() `+`(<textstat>,<textstat>) subset(<textstat>) as.data.table(<textstat>) show(<textstat>) p_attributes(<textstat>) knit_print(<textstat>) get_corpus(<textstat>) format(<textstat>) restore() cp() view(<textstat>)

S4 textstat superclass.

Methods III: Fulltext Output

A key consideration of polmineR is to combine quantitative and qualitative steps of analysis in a workflow. A set of methods is exposed to generate and enhance fulltext output.

read()

Display full text.

html()

Generate html from object.

annotations() `annotations<-`() edit(<textstat>)

Annotation functionality

as.markdown()

Get markdown-formatted full text of a partition.

mail() browse()

Defunct functionality

view()

Inspect object using View().

highlight()

Highlight tokens in text output.

tooltips()

Add tooltips to text output.

get_template()

Get template for formatting full text output.

get_type()

Get corpus/partition type.

partition_to_string

Decode as String.

The ‘textstat’ Class & Childs

The main feature of the textstat class and of classes inheriting from it is that it keeps statistical information about a corpus or partition in a data.table in the field ‘stat’. Methods defined for the textstat superclass are available for its child classes, unless they are overloaded in a specified manner.

name(<textstat>) name(<character>) `name<-`(<textstat>) round(<textstat>) sort(<textstat>) as.bundle() `+`(<textstat>,<textstat>) subset(<textstat>) as.data.table(<textstat>) show(<textstat>) p_attributes(<textstat>) knit_print(<textstat>) get_corpus(<textstat>) format(<textstat>) restore() cp() view(<textstat>)

S4 textstat superclass.

summary(<count>) length(<count>) hist(<count>)

Count class.

ngrams_class ngrams-class

Ngrams class.

p_attributes(<partition>) p_attributes(<subcorpus>) is.partition() enrich(<partition>) as.regions(<partition>) split(<partition>)

Partition class and methods.

sample(<hits>)

S4 class to represent hits for queries.

as.data.table(<ranges>)

Ranges of query matches.

length(<context>) p_attributes(<context>) count(<context>) sample(<context>) enrich(<context>) as.regions(<context>) trim(<context>)

Context class.

show(<cooccurrences>) as.data.frame(<cooccurrences_bundle>) format(<cooccurrences>) view(<cooccurrences>) view(<cooccurrences_reshaped>)

Cooccurrences class.

as.simple_triplet_matrix(<Cooccurrences>) as_igraph(<Cooccurrences>) subset(<Cooccurrences>) decode(<Cooccurrences>) kwic(<Cooccurrences>) as.sparseMatrix(<Cooccurrences>) enrich(<Cooccurrences>)

Cooccurrences class for corpus/partition.

Cooccurrences(<corpus>) Cooccurrences(<character>) Cooccurrences(<slice>) Cooccurrences(<partition>) Cooccurrences(<subcorpus>)

Get all cooccurrences in corpus/partition.

summary(<features>) show(<features>) summary(<features_bundle>) format(<features>) view(<features>)

Feature selection by comparison.

The ‘kwic’ & ‘Labels’ Classes

xxx

get_corpus(<kwic>) count(<kwic>) as.DocumentTermMatrix(<kwic>) as.TermDocumentMatrix(<kwic>) show(<kwic>) knit_print(<kwic>) as.character(<kwic>) `[`(<kwic>,<ANY>,<ANY>,<ANY>) subset(<kwic>) as.data.frame(<kwic>) length(<kwic>) sample(<kwic>) merge(<kwic_bundle>) enrich(<kwic>) format(<kwic>) view(<kwic>)

S4 kwic class

The ‘bundle’ Class & Childs

S4 classes for managing text analysis.

`name<-`(<bundle>) length(<bundle>) names(<bundle>) `names<-`(<bundle>,<vector>) unique(<bundle>) `+`(<bundle>,<bundle>) `+`(<bundle>,<textstat>) `[[`(<bundle>) `[`(<bundle>,<ANY>,<ANY>,<ANY>) `[[<-`(<bundle>) `$`(<bundle>) `$<-`(<bundle>) sample(<bundle>) as.bundle(<list>) as.bundle(<textstat>) as.data.table(<bundle>) as.matrix(<bundle>) subset(<bundle>) as.list(<bundle>) get_corpus(<bundle>)

Bundle Class

show(<partition_bundle>) summary(<partition_bundle>) merge(<partition_bundle>) barplot(<partition_bundle>) as.partition_bundle(<list>) partition_bundle(<environment>) enrich(<partition_bundle>) enrich(<subcorpus_bundle>) flatten()

Bundle of partitions (partition_bundle class).

context_bundle-class show,context_bundle-method summary,context_bundle-method [,context_bundle-method [,context_bundle,ANY,ANY,ANY-method [[,context_bundle-method

S4 context_bundle class

blapply()

apply a function over a list or bundle

Low-level CWB Access

The API used to access CWB indexed corpora is the RcppCWB package. A set of functions in the polmineR package serves as an intermediate interface to access corpus data.

registry_move() registry() data_dir()

Get registry and data directories.

registry_reset()

Reset registry directory.

registry_get_name() registry_get_id() registry_get_home() registry_get_info() registry_get_encoding() registry_get_p_attributes() registry_get_s_attributes() registry_get_properties()

Evaluate registry file.

cpos cpos,corpus-method cpos,character-method cpos,slice-method cpos,partition-method cpos,subcorpus-method cpos,matrix-method cpos,hits-method cpos,NULL-method

Get corpus positions for a query or queries.

get_token_stream()

Get Token Stream.

decode()

Decode corpus or subcorpus.

is.cqp() check_cqp_query() as.cqp()

Tools for CQP queries.

Utility Functions

Exported utility functions / helpers.

as.utf8() as.nativeEnc() as.corpusEnc()

Conversion between corpus and native encoding.

capitalize()

Capitalize character vector.

href()

Add hypertext reference to html document.

Backwards Compatibility

Starting with v0.7.9, the coding style of the polmineR-package has moved from camelCase to snake_case. A set of ‘old style’-functions ensures backwards compatibility and that old code will not break.

sAttributes() pAttributes() getTokenStream() getTerms() getEncoding() partitionBundle() as.partitionBundle() corpus(<textstat>) corpus(<bundle>) corpus(<kwic>)

Renamed Functions